The study aims are: a) to measure predictor variables at two time periods to study the inter-relationship among psychosocial predictor variables early and late in the pregnancy and the manner in which they are predictive of complications of pregnancy; 2) to extend findings of a previous study with middle class women to a lower socio-economic population of Black, Caucasian, and Hispanic women; and c) to obtain detailed information about the types and sources of social support that are most effective in buffering the deleterious effects of stress on pregnancy outcomes. A battery of five standardized instruments will be administered to 350 medically-normal women before 18 weeks gestation and again between 32-38 weeks gestation. Based on previous work, the variables of life stress, social support, and emotional disequilibrium (anxiety, depression, and low self-esteem) are expected to be predictive of complications of pregnancy. The outcome variables are complications of gestation, complications of labor and delivery, and complications related to the infant's condition (including gestational age and birth weight). These complications will be determined by blind chart review after the pregnancy has ended. The long-term objectives of this study are: a) to refine criteria for determining high risk on the basis of psychosocial status for a lower SES population of ethnically diverse pregnant women; and b) to establish the necessary baseline data for planning an intervention to augment key types and/or sources of social support to reduce complications of pregnancy in a subsequent study.